Monday 31 March 2014

Minority languages and court proceedings: the possibilities and limits of EU law

Steve Peers

In its recent judgment in Ruffer, the Court of Justice of the European Union (CJEU) confirmed and extended its case law relating to the use of minority languages in court proceedings.

The case concerned Mrs. Ruffer, a German national living in Germany, who was injured in a skiing accident in Bolzano, a part of Italy where German is widely spoken. She blamed Ms. Pokoma, a Czech national resident in the Czech Republic, for the accident, and sued her in the Italian courts, presumably because (in accordance with EU rules on civil jurisdiction) Italy was the place where the harmful event occurred. Mrs. Ruffer used German as the language of proceedings and Ms. Pokoma did not object to this.

While Italian law permits German to be used as the language of proceedings in Bolzano, a higher national court ruled in the meantime that only Italian citizens residing in Bolzano could invoke this right. So the civil proceedings were void. The referring national court asked the CJEU whether this was correct in light of EU law, particularly considering the Court’s previous judgment in Bickel and Franz, where it had ruled that Austrian citizens merely travelling through Bolzano could rely, as EU citizens, upon the right to non-discrimination on grounds of nationality in order to insist upon the use of German in criminal proceedings.

 According to the CJEU, the Bickel and Franz judgment applied also to civil proceedings, given that otherwise there would be discrimination between Italian citizens living in Bolzano, on the one hand, and the citizens of other Member States who can speak German, on the other hand. The Court rejected the Italian government’s argument that the national courts could not manage such an extension of language rights, on the grounds that the referring national court had asserted just the opposite. Equally it rejected the Italian government’s argument that greater use of German would lead to greater costs, because this was not a permissible argument in the context of EU citizenship.


This judgment is not particularly surprising in light of the Court’s prior ruling in Bickel and Franz, but its implications and context are worth some further discussion.

First of all, it should be emphasised that the EU’s non-discrimination rule as regards minority languages is dependent upon the treatment which a Member State accords to those languages in the first place: in this context, EU law guarantees non-discrimination on grounds of nationality, not the underlying minority language rights as such. Whether a Member State extends such rights in the first place is in principle determined by its national law and international commitments, not EU law, subject to the possible argument that discrimination against minority languages could amount to indirect discrimination on grounds of racial and ethnic origin, banned by the EU’s Race Discrimination Directive.

Secondly, the EU law rule cuts across the concept of a ‘national minority’ as defined in international treaties on minority rights. The parties to the civil proceedings are neither Italian nationals nor (in their own countries) part of a minority language group (presuming that Ms. Pokoma speaks German as a second language).

Thirdly, the language rights in these cases go further than the rights guaranteed by Article 6(3)(e) ECHR, as applied by the EU Directive on interpretation and translation in criminal proceedings, both because they concern civil (rather than criminal proceedings) and because they concern the main language of the proceedings, not merely a right to interpretation and translation.

Fourthly, the Court’s unwillingness to limit its prior judgments to criminal proceedings only suggests that any rights (not just rights related to any form of judicial proceeding) which a Member State grants to its language minorities must in principle be granted to the nationals of other Member States – even where the language concerned is (as in Ms. Ruffer’s case) the majority language of that person’s state of origin. Perhaps, however, the arguments concerning administrative inconvenience might be better evidenced in another case.

 Fifthly, the right does not depend upon residence in the Member State concerned. Logically even if one or both of the parties had never set foot in Bolzano, but the EU civil jurisdiction rules required proceedings to take place there nonetheless, the equal treatment rule would still apply. However, as regards other aspects of minority language rights, such as education and communication with the administration, it might be arguable that equal treatment would depend (as it does in some other CJEU judgments) upon having achieved a sufficient degree of integration into the territory of the Member State concerned (or the relevant part of it), at least in the form of residence there.

Finally, while the CJEU rejected the costs argument out of hand from a legal perspective, it is certainly possible that this argument could be relevant politically. National or regional politicians might use this line of judgments from the CJEU to argue that it is too expensive to extend or retain minority language rights, because of the consequence of extending them to nationals of other Member States. This would be an unfortunate and unintended consequence of the Court’s rulings.

Barnard & Peers: chapter 13

Friday 28 March 2014

The EU’s Returns Directive: Does it improve or worsen the lives of irregular migrants?

Steve Peers

Many EU citizens are concerned about the social and economic impact of immigration, particularly irregular (sometimes called ‘illegal’) migration of non-EU citizens (third-country nationals) who do not have a right to enter or stay on the territory of an EU Member State. However, others are more concerned to ensure that irregular migrants are treated humanely and decently. To control and regulate irregular migration, the EU has adopted a number of measures, the most prominent of which is known as the ‘Returns Directive’. This Directive, adopted in 2008, governs a broad range of issues, in particular an obligation to return irregular migrants, their treatment during expulsion proceedings, entry bans, procedural rights and the grounds and conditions for detention.

When the Returns Directive was adopted, it was controversial among NGOs and the academic world, because of a perception that it took an unduly harsh approach on these issues. In the five years since its adoption, the Directive has been the subject of much litigation before the Court of Justice of the European Union (CJEU), largely (but not only) as regards its detention rules. [Update: for more on this CJEU case law, see my 2015 paper in the European Journal of Migration and Law].  We now have the opportunity to consider the impact of the Directive more fully, in light of today’s report by the European Commission on its application. This report forms part of a broader reflection by the Commission on EU expulsion policy. The Commission’s analysis of the implementation of the Directive, and this reflection on the broader aspects of returns policy, should be considered in turn.

Implementing the Returns Directive

Member States had to implement the Directive by Christmas Eve 2010, and the European Commission is required to report on its implementation every three years. The Directive does not apply to the UK and Ireland, but to the extent that it applies to those who enter the territory without authorisation, it applies to Denmark and the Schengen associates (Norway, Iceland, Switzerland and Liechtenstein).

Interestingly, according to the report, the Commission has already embarked upon a programme to ensure the correct implementation of the Directive in the Member States. The report summarises the results of this programme, by indicating how many Member States still needed to change their law after the transposition deadline in order to comply with key rules in the Directive, and how many of those States have implemented those changes.

In particular:

 - 6 out of 11 Member States have clarified the concept of ‘risk of absconding’ (an important issue as regards grounds for detention);

- 6 out of 7 Member States have changed their law to make clear that detention is not justified if there is no reasonable prospect of removal;

 - 4 out of 6 Member States now admit NGOs and international bodies to detention centres;

 - 4 out of 6 Member States now provide for free legal assistance to irregular migrants, subject to the conditions set out in the Directive;

- 13 out of 16 Member States have set up a forced return monitoring system;

 - 11 out of 14 Member States now take account of the EU guidelines on forced returns by air.

 The Commission then examines a number of specific legal issues in detail. As regards detention, where the Directive requires a review of detention after a ‘reasonable’ period, there is a wide variation between Member States, with some requiring frequent reviews and some providing only for a review at the end of the initial six-month detention period. A six-month wait for a review of detention cannot plausibly be considered to be a ‘reasonable’ period, and moreover frustrates the intention of providing for a review in the first place, since in principle detention can only be justified for six months, with a further extension of up to 12 months only on limited grounds. However, the Commission does not comment on this issue, or promise to challenge the Member State(s) in question. The best interpretation of the ‘reasonable’ review period is that, in the absence of frequent ex officio reviews, detention must be reviewed whenever the migrant presents a prima facie case that the grounds for detention are no longer satisfied.

It would be useful to know how often appeals against detention are successful and on what grounds, but the Commission does not discuss this issue. The Commission admits that it does not have statistics on the grounds for which irregular migrants are detained.

Next, the Commission presents details of the development of alternatives to detention in a number of Member States, although it does not present empirical evidence of how often these alternatives are applied in practice, as compared to how many people are detained, and how the Directive has changed the position in practice.

The Commission refers to some good practice as to the treatment of irregular migrants who cannot be removed, but the Directive does not address this issue in detail.

As for the maximum length of detention, the Commission indicates that 12 Member States reduced their maximum period to the 18-month maximum in the Directive after it was adopted, while another 8 Member States increased their detention period up to 18 months. Five Member States left their detention periods unchanged, while another three have apparently provided for detention where it did not exist previously (or perhaps the Commission simply lacks data on the prior rules applied in those Member States).

As the Commission points out, the maximum periods of detention are not usually applied, and it supplies data on how long irregular migrants are actually detained in practice to prove this point. But in the absence of data about how long irregular migrants were detained for in practice before the Directive was adopted, it is impossible to be sure what effect it has had on the actual length that migrants spent in detention.

Oddly, the Commission report says nothing about the rules governing the possible extended period of detention for up to 18 months. Since these extended periods can only be justified on a limited number of grounds, the Commission should have reported on whether Member States have properly complied with the relevant rules and how many people are detained for longer periods in practice.

The Commission rightly notes that inhumane detention conditions for irregular migrants fall within the scope of EU law, given the regulation of immigration detention by the Directive, and threatens to sue Member States on this point. But in light of the cardinal importance of the ban on torture or inhuman or degrading treatment, the Commission should surely feel a greater sense of urgency about this issue.

Next, the Directive requires that in principle immigration detainees cannot be detained in prisons. If it is not possible to apply this rule, then those immigration detainees being detained in prisons must be detained separately from other prisoners. The Commission reports that half of Member States do detain irregular migrants in prisons, and nine Member States do not comply with their obligations to keep immigration detainees separate from ordinary prisoners. It threatens infringement actions against these States. Moreover, as the Commission notes, there are pending cases before the CJEU addressing these issues (in fact, the cases which the Commission refers to will be heard by the CJEU before Easter).

As for other rules on detention conditions, the Commission reports that Member States have correctly legislated as regards detainees’ health care and communication with the outside world. But it reports that there are doubts about the correct application of these rules in practice, and promises follow-up. Similarly, Member States have achieved de jure, but arguably not de facto, compliance with the Directive’s rules on the detention of families and children. Here the Commission does not expressly promise to follow the issue up, even though it reports that significant numbers of children are being detained, and although the Directive states that children can only be detained as a ‘last resort’. Again, there are no statistics on the detention of children, and the effect of the Directive on this issue cannot be judged.

The Commission begins its assessment of the other provisions of the Directive by examining the rules on voluntary departure. Member States are obliged to offer irregular migrants the possibility of voluntary departure, with only limited exceptions. On this issue there have been many positive changes in national law, although the Commission does not assess what is actually happening in practice. It admits that it does not have statistics concerning voluntary departure, although it quotes a Frontex report indicating that 44% of removals in 2012 were voluntary, whereas 56% were forced. Without seeing the evolution of these statistics over time, the impact of the Directive on this point cannot be assessed.

Then the Commission moves on to consider the rules on monitoring of forced removals. Again, it notes positive legislative changes in many Member States, and threatens infringement actions against Member States which have not complied with the rules, but does not consider the practical impact. More precisely, while the report gives specific details as regards the joint return operations arranged by Frontex, the EU border agency (see below), it does not provide similar details as regards the percentage of returns by Member States which are actually monitored, and as to whether any complaints have been made.

Nor does the Commission assess: whether Member States have complied with their obligation to postpone removal in specified cases; how well Member States have complied with their overarching obligation to take account of the best interests of the child, family life, non-refoulement and the state of health of migrants when applying the Directive; and in particular whether removal operations have been ‘proportionate’, used only ‘reasonable force’, were consistent with ‘fundamental rights’ and observed the ‘dignity’ and ‘physical integrity’ of irregular migrants.

According to the report, most Member States have exercised the options to exclude from the scope of the Directive those being expelled due to a criminal offence, and those who were apprehended attempting irregular entry. However, the Commission does not assess whether Member States have interpreted these exceptions correctly in light of the CJEU’s case law, which makes clear that the ‘criminal law’ exception cannot apply simply because irregular entry or residence is a criminal offence in a Member State.

Even when irregular migrants are excluded from the scope of the Directive because they were apprehended attempting irregular entry, the Directive nevertheless requires that Member States apply some key rules to protect the persons concerned, as regards the conduct of removals, detention conditions, non-refoulement and emergency health care. The report states that these rules were applied in the ‘majority’ of cases (without further clarification), but appears rather insouciant about the cases where the rules were not applied.

As for the procedural rights of irregular migrants, there is again de jure compliance but some de facto non-compliance in practice, which the Commission largely does not promise to follow up on, except as regards the rules on legal aid. The majority of Member States do not provide for automatic suspensive effect of appeals.

On the important question of whether irregular migration can be criminalised, the Commission correctly notes that this issue is not directly regulated by the Directive, but provides very useful maps of which Member States criminalise irregular entry and/or residence, and by which means (fines or custodial sentences). As the Commission points out, the case law of the CJEU has curtailed Member States’ ability to impose custodial sentences for irregular migration, on the grounds that this interferes with the effectiveness of the removal process. It states that infringement procedures are underway to ensure the correct application of this case law by Member States.

Next, as regards the obligation in principle to expel irregular migrants set out in the Directive, the Commission states that this has had no impact on the numbers of apprehensions of irregular migrants. But one could hardly expect it to do so, since the rule does not concern apprehension as such. The key issue is whether the Directive has affected the numbers of irregular migrants who are subjected to removal orders after their apprehension, and the number of cases where these removal orders are carried out. While the Commission provides overall statistics on the numbers of apprehensions, removal orders, and removals carried out from 2010-2012, it makes no attempt to link these statistics to the application of the Directive.

Finally, as regards the rules on entry bans, the report indicates a convergence around the five-year maximum entry ban period which applies to most cases, with six Member States increasing the number of entry bans and eight Member States reducing the maximum time limits for such bans. Again, detailed statistics on the use of such bans in practice are not available, and the Commission does not assess Member States’ application of the detailed exceptions to the general rules in much detail.

Broader aspects of expulsion policy

The Commission’s assessment of the broader aspects of expulsion policy begins with an analysis of the role of Frontex, the EU’s border control agency, as regards coordinating joint expulsions. According to the report, Frontex arranges the joint expulsions of about 2000 irregular migrants a year, and about half of these joint expulsions are monitored, with no complaints in practice. The Commission urges Frontex to ensure that all joint expulsions are monitored, although its position is weakened by its questionable suggestion that Frontex is not obliged to do this. It refers to the appointment of a Frontex Fundamental Rights Officer, who has competence to examine the compliance of Frontex joint operations with fundamental rights, but does not assess how well this task is being carried out in practice. In particular, the Commission does not refer to the recent recommendation of the European Ombudsman that Frontex consider complaints from individuals affected by its activities.

As for future developments, the Commission intends to draw up a non-binding Returns Handbook regarding the implementation of the Directive. It also endorses a study on alternatives to detention being drawn up by the European Migration Network. It will consider proposing to legislate for an obligation to issue alerts in the Schengen Information System every time a Member State issues an entry ban, when it reviews that System in 2016, although it states that this already happens in practice in every case. The Commission also encourages the European Migration Network to draw up a list of best practices in the operation of returns procedures, supports the codification of Council of Europe rules on detention conditions and seeks to continue (without any concrete details) cooperation and dialogue with third States on returns issues. On the latter point, there is a stark lack of transparency as regards Member States’ arrangements with third countries, and about what happens to returned persons in practice. Finally, the Commission plans the collect information on best practice as regards the situation of people who cannot be removed, although it is not clear what it plans to do with this information.


Has the Returns directive improved or worsened the position of irregular migrants? When it was first adopted, the Directive was widely seen in the NGO community and among academics as a highly negative measure that offered little protection for irregular migrants, but rather encouraged Member States only to lower their standards. In light of the case law of the CJEU, and the information on national implementation of the Directive set out in this report, this assessment has to be more qualified.

It is clear that, as regards the length of detention and the use of entry bans, standards have been raised in some Member States and lowered in others. While it is always possible that the latter group of Member States would have lowered their standards on their own initiative, it cannot be ignored that they lowered those standards at the time of implementing the Directive, and that when they did so, they matched the Directive’s rules. This can hardly be a coincidence.

While the Commission’s efforts to ensure the correct implementation of the Directive are useful, they were also clearly rather tardy, and threatening to begin infringement proceedings only in 2014 is rather sluggish. As regards detention issues in particular, Article 5 ECHR requires all aspects of detention to be in accordance with the law, and this must logically refer to EU law as well as national law. So ensuring the correct enforcement of EU law as regards detention is particularly important.

Although the plan to draw up a guidance document as regards the Directive is welcome, the Commission could and should have done this earlier. If it had drawn up such guidance even before the Directive’s implementation deadline, rather than over three years afterwards, it could have avoided some errors in transposition. For instance, it recently drew up a guidance document as regards the implementation of the EU’s crime victims’ Directive, even though Member States do not have to apply that law until the autumn of 2015.

In fact, I can prove from personal experience that the Commission could have produced a guidance document on the Returns Directive before the implementation deadline, because I co-authored a lengthy report (with Olivier de Schutter of Leuven University) for the EU’s Fundamental Rights Agency back in 2009, on the correct interpretation of the Directive in light of international norms. Of course, it is hard for me to believe that the Commission could have disagreed with our interpretation! But even if it did so, it could at least have used our analysis as a source and a starting point for drawing up guidelines.

For the Returns Directive, the Commission’s tardiness matters less than it might, because of the significant role of the CJEU in practice as regards references from national courts concerning the Directive. In fact, this law has attracted more references to the CJEU than most EU immigration or asylum measures. While the Court’s case law has on the whole interpreted the Directive more liberally than its wording might suggest, it has focus more on the objective of efficient expulsion, rather than on irregular migrants’ human rights.

It is clear from the Commission’s report on this Directive that the usual dichotomy between law on the books and the practice of law on the ground is particularly sharp as regards this Directive. This cannot be easily solved by the usual means, so perhaps some fresh thinking is required. Building on the existing commitment to provide for an effective forced-return monitoring system, one way forward might be to expand on this system to require each Member State to provide for an effective supervision and complaints process (without prejudice to the role of the courts) as regards the implementation of this Directive.

The Directive required the Commission’s first implementation report to examine in particular the basic rules on detention, the provision on entry bans and the effect of the legal aid requirements upon Member States. Its report does examine the first two of these issues, but not the third. More broadly, the Commission does not expressly examine whether it ought to propose any amendments to the Directive.

This report raises implicit questions about the process of adopting implementation reports in areas of EU law, such as irregular migration, which impact greatly upon human rights. While this particular report does mention human rights issues, including inhumane detention conditions, it is arguable that such reports should always include a specific review of Member States’ compliance with the EU Charter of Fundamental Rights when applying the law, and also a consideration of whether the rights in the Charter could be promoted by amendments to the relevant legislation.

Overall, it's not possible to assess whether the Directive has positively or negatively affected irregular migrants without substantially more information about its application in practice. What we can conclude is that its correct and timely application has been hindered by the Commission’s initially cautious approach, and that we need to consider new ways of ensuring compliance with EU law in areas such as these.

 Barnard & Peers: chapter 26

Family Reunion for EU Citizens: The CJEU clarifies the Singh and Carpenter judgments

Chiara Berneri, Lecturer at BPP Law School

When can an EU citizen who lives in his or her own Member State claim a right under EU law to be joined by his or her third-country national family members? Traditionally EU law cannot be invoked at all in such cases, but there are important exceptions from that rule. The Court of Justice ruled on the two most important exceptions in two separate judgments of 12 March, following the December 2013 opinion of Advocate-General Sharpston (see the previous blog post on this opinion). Unfortunately, these judgments contained some unpleasant surprises.

Clarifying the Singh judgment: the O and B case

In 2007 O (a third-country national) and sponsor O (ie the Dutch citizen related to O) took residence in Spain. In reality, two months after arriving in Spain, in 2007, sponsor O returned to the Netherlands because she could not find a job there. However, since her departure, she continuously flew to Spain over weekends and enjoyed services there. Since 1 July 2010 O has been registered as residing with sponsor O in the Netherlands. However, his application for a document showing lawful residence was refused.

B is a Moroccan national who lived in the Netherlands, since 2002, with sponsor B (a Dutch citizen). After being sentenced to two months’ imprisonment for using a false passport he invoked the ‘Belgian route’ well-known to Dutch immigration lawyers, moving to Belgium where sponsor B rented out a flat for them. Since sponsor B was not able to find a job in Belgium she moved back to the Netherlands but came back every weekend to spend time with B. The couple got married in Morocco in 2007. In 2009 the Netherlands lifted the declaration of undesirability against B. The couple moved to the Netherlands. B applied for a document showing his lawful residence but it was refused.

The Court’s judgment

The Court starts its reasoning in the O and B judgment by pointing out the essence of the question asked by the referring court: whether Directive 2004/38 (the EU citizenship Directive) and Art. 21 TFEU (the Treaty rule on EU citizens’ right to move and reside freely) must be interpreted as precluding a Member State from refusing a right of residence to a third country national who is a family member of a Union citizen holding the nationality of that Member State, following the return of that citizen to that Member State, in circumstances in which the Union citizen resided with the family member in another Member State solely by virtue of being a EU citizen. The Court, recalling its previous judgments, began by stating that neither Directive 2004/38 nor Art. 21(1) TFEU confer any autonomous right on third country nationals but, on the contrary, any rights conferred on third country nationals by provisions of EU law concerning Union citizenship are rights derived from the exercise of freedom of movement by a Union citizen.

With regard to Directive 2004/38, the Court pointed out that the subject of the directive is the conditions governing the exercise of the right of movement and residence of EU citizens and their family members in another Member State. Since, according to international law, a state cannot refuse entry and residence to its own nationals, it follows that Directive 2004/38 is intended only to govern the right of entry and residence of a Union citizen in a Member State other than the Member State of which he or she is a national. Hence, it follows that this Directive does also not intend to confer a derived right of residence on third country national family members of such EU citizens, unless those citizens have moved to another Member State.

At this point, the Court starts wondering whether a derived right of residence may, under some circumstances, be based upon Art. 21(1) TFEU. The Court recalls the well known deterrence doctrine by stating that the refusal to allow the right of residence to the third country national family members of an EU citizen who returns to his or her Member State of nationality after exercising free movement rights in another Member State (a ‘returnee’) would be such as to discourage that Union citizen from exercising his or her right of movement. It then mentions the previous Singh and Eind cases, in which it had ruled that an obstacle to leaving the Member State of which the worker or self-employed person is a national is created by the refusal to confer, when the worker or self-employed person returns to his or her Member State of origin, a derivative right of residence on the family members of that worker or self-employed person who are third country nationals.

Can then Singh and Eind be applied also to this case? The Court answers affirmatively. In fact, according to the Court, the grant of a derived right of residence to a third country national family member of a EU citizen who has resided in another Member State solely by virtue of his or her being a Union citizen, seeks to remove the same kind of obstacles that a worker or self-employed person (such as in Singh and Eind) leaving for another Member State would find. 

The Court interestingly also specifies that the conditions of residence granted to the third country national family member should not be stricter than those provided by Directive 2004/38. This Directive in fact should be applied by analogy to returnees’ third-country national family members. This statement leads the Court to make some interesting considerations, which characterize the novelty of this case.

The Court states that an obstacle to free movement and residence arises only when the residence of the Union citizen in the host Member State has been sufficiently genuine. Since, as mentioned before, Directive 2004/38 applies by analogy, the Court describes the meaning of “sufficiently genuine” by referring to that Directive. According to the Court, if someone moves to another Member State pursuant to Art. 6(1) Directive 2004/38 does not intend to settle there (since Art. 6 concerns just the right to stay of the EU citizen in another Member State up until 3 months). Hence, in such a case, the denial of the third country national’s right of residence in the Member State of origin would not deter the EU citizen from exercising his/her right of free movement. On the other hand, if the Union citizen intends to exercise his or her right to reside for more than three months in another Member State pursuant to Art. 7(1) and (2) of Directive 2004/38 the potential denial of the right of residence to the third country national family member once back in the state of origin is likely to deter the citizen from leaving the state of origin in the first place. In fact if, in conformity with the conditions set out in Art. 7(1) and (2), family life is created and strengthened in the host Member State, the Union citizen has to be able to keep on enjoying the same conditions of family life once he or she returns to his or her Member State of nationality.

  A fortiori, according to the Court, the same logic will apply in case the Union citizen and the family member were granted a permanent right of residence in the host Member State in accordance with Article 16(1) and (2) of Directive 2004/38. Interestingly, the Court concludes that it is up to the referring court to determine whether sponsor O and B settled and genuinely resided in the host Member State. It also adds that the scope of Union law cannot be extended to cover abuses.

Clarifying Carpenter: the S and G case

S is a Ukrainian national. Her Dutch son-in-law, sponsor S, has worked since 2002 for an employer established in the Netherlands but spends 30% of his time preparing and making business trips to Belgium, to which he goes at least once a week. S looks after her son-in-law’s child. S applied for a document certifying lawful residence in the Netherlands but it was refused. 

G is a Peruvian national. She married sponsor G in Peru. Sponsor G lives in the Netherlands but works for a Belgian employer and he travels daily to Belgium. G’s application for a document certifying lawful residence in the Netherlands was rejected.

The Court’s judgment

In essence, the main question that the Court asks in its judgment is whether Directive 2004/38, Art. 20 TFEU (the basic rule on EU citizenship), Art. 21 TFEU and Art. 45 TFEU (the basic rule on free movement of workers) must be interpreted as precluding a refusal by a Member State to grant family residence rights to a third country national family member where the EU sponsor is a national of the Member State and resides regularly in that Member State but regularly travels to another Member State in the course of his professional activities. The Court initially refers to the previously mentioned case of O and B and confirms that Directive 2004/38 does not confer the derived right of residence on third country nationals who are family members of a Union citizen in the Member State of which that citizen is a national.

Next, the Court examines whether Art 45 TFEU could be applicable. Since the referring court cited the Carpenter case (where the CJEU held that an EU citizen providing services in another Member State might be able to rely on the Treaty rules on free movement of services in Article 56 TFEU in order to secure the entry and residence of a third-country national family member in his or her State of nationality) for this purpose the Court clarifies the scope of that judgment. In the CJEU’s view, the circumstances of Union citizens such as S and G do fall within the scope of Art. 45. Moreover, the Court confirms that its interpretation of Art. 56 TFEU in Carpenter is transposable to Art. 45 TFEU.

The Court then refers again to the deterrence approach by stating that a derived right of residence is based on the fact that a refusal to allow it would be such as to interfere with the exercise of fundamental freedoms guaranteed by the TFEU. Finally, it concludes that it is up to the national court to determine whether, in the circumstances of this case, the grant of a derived right of residence is necessary to guarantee the citizen’s effective exercise of the fundamental freedom guaranteed by Art. 45 TFEU. However, rather cryptically, it also adds that the mere fact that it might appear desirable that the child is taken care of by the Union citizen’s ascendant, it is not sufficient in itself to constitute a dissuasive effect.


The solutions proposed in both judgments are particularly controversial, especially when compared to A.G. Sharpston’s opinion. With regard to the case of O and B (the post-Singh case), the Court offers a completely different analysis than the one suggested by A-G Sharpston. The A-G did not agree with the idea that the EU citizen should reside in the host Member State for a certain amount of time before he or she can claim derived residence rights for his or her third country national family member, suggesting instead that the answer (whether the third country national should be allowed residence in the national Member State of his/her EU relative) depends on why the EU citizen and his/her family member(s) were not moving together. She pointed out that EU citizens enjoy the freedom to decide themselves how to exercise the right to a family life. The fact that some prefer to live with their family members and others might, at a particular moment, have other priorities or face difficulties to live together immediately does not mean that that they should not be allowed to enjoy the right of residence in the national Member State of the EU citizen. The only circumstance in which she seemed to suggest that the right of residence should not be granted is when a third country national family member and an EU citizen have decided that they no longer wished to live together as a couple.

However, the Court suggests that an obstacle to free movement and residence arises only when the residence of the Union citizen in the host Member State has been sufficiently genuine (i.e. residence pursued for more than three months, in accordance with Art. 7 Directive 2004/38). Surely, it is true that the test suggested by the Court seems to provide more clarity than the solution proposed by A-G Sharpston. Indeed, a test based on the lapse of time spent together in the host Member State by the family unit is more easily applicable. Nevertheless, it is also true that, by looking at the concrete application of the tests to the circumstances at stake (or at least at the suggestions that have been given to national courts), it seems that A-G Sharpston’s solution is potentially able to cover more situations.

The use of the test endorsed by the Court, most likely, will be applied by the Dutch national court as not granting, both to O and B, a right of residence in the Netherlands. Indeed sponsor O resided just for two months in Spain and sponsor B simply visited B at week-ends (at para 59 the Court rules out the possibility that short periods of residence have a cumulative effect of creating a derived right of residence for a family member of a Union citizen). On the other hand, A-G Sharpston, on the basis of her test, suggests that O should be granted a right of residence in the Netherlands (subject to the conditions and limitations of Directive 2004/38). With regard to B, A-G Sharpston instead suggests that he should not be granted residence in the Netherlands but simply because sponsor B is a partner and not an official family member (spouse) and, therefore, does not fall within the scope of Directive 2004/38.

The decision of the Court in O and B seems also to partially resonate with the new returnees rules inserted in the recent UK regulations 2013 [The Immigration (European Economic Area) (Amendment) (No. 2) Regulations 2013], which require a British citizen to have transferred the centre of his/her life to another Member State in order to acquire a right of residence in the UK for his/her third country national family member seeking a right to reside in the UK upon their return. It is true that the judgment is not as strict as the rules themselves, which talk about transferring the centre of one’s life in the host Member State, whereas the Court just refers to the amount of months the EU citizen has to be there in order to claim family reunification (for a full discussion of the impact of this judgment on the UK rules, see the post in the UK free movement blog). However, it is also true that the requirement of having resided in the host Member State for the purpose of Art. 7 of Directive 2004/38 suggests a certain degree of integration and engagement in the host State that aligns the case more towards these new rules and detaches it more clearly from the solution proposed by A-G Sharpston.

This is even more evident if we consider that the Court points out that the scope of Union law cannot be extended as to cover abuses and that an abuse of Union law consists in “the intention to obtain an advantage from the European Union rules by artificially creating the conditions laid down for obtaining it”. If this extra consideration will be applied in future rulings, it is likely that the Court will end up granting third country national residence rights in the Member State of origin of the EU citizen just when the latter decided to move to another Member State for reasons that do not encompass taking advantage of the more friendly EU family reunification rules. Once again, this is quite different from what A-G Sharpston suggested in her opinion, in which she stated that the real issue is not the abuse of EU law but the free choice of the EU citizen to decide whether to move or not to move. A national measure that imposes movement restricts that choice and is contrary to Art. 21(1) TFEU.

As for the S and G case, the position of the Court again seems to be much stricter that the one endorsed by the Advocate-General. In relation to the specific circumstances concerning S, A.G. Sharpston suggests that, while evaluating the presence of a causal link with EU law, the referring Court should examine whether denying residence to S would cause sponsor S to seek alternative employment that would not involve the exercise of right of free movement or cause him to move with his family, including S, to another Member State. The Court, on the other hand, instead of indicating the elements that the referring court should take into account in order to achieve a fair solution, stated that the simple fact that it is desirable that the child is looked after by the third country national grandmother is not sufficient in itself to constitute a dissuasive effect to the free movement rights exercised by sponsor S. It would be interesting to find out on which grounds the Court dared to make such a bold statement which, furthermore, is not even supported by any concrete evidence.

Indeed, in Carpenter the Court at least pointed out how the circumstances of the case suggested that the separation of Mr and Mrs Carpenter would have been detrimental to their family life and to the conditions under which Mr Carpenter exercised his fundamental freedom because Mrs. Carpenter looked after Mr. Carpenter’s children and allowed him to pursue his business. In the circumstances of B no reference to the concrete facts of the case is made. Why was the presence of Mrs Carpenter in the UK essential and now the presence of O in the Netherlands superfluous? The Court, implicitly, seems to indicate a distinction based on different categories of relatives. If this is really the case, it is a pity that the Court missed an opportunity to specify why this difference should be in place. Moreover, if really discretion has to be given to national courts, is it fair to muddy the waters with this consideration?

 Finally, unlike the O and B case, the Court in S and G did not find that Directive 2004/38 should apply by analogy but instead relied on Art. 45 TFEU. This means that, most likely, third country national family members will not be able to rely on the Directive conditions with regard to access to employment, social benefits etc. Hence, supposing that the national Dutch court will grant residence rights to S and G (apparently, this will be very unlikely in the case of S since the Court stated that the mere fact that it may appear desirable for the child to be cared for by the third country national this is not sufficient in itself to constitute a dissuasive effect to free movement), under which conditions will they be able to reside in the Netherlands? Will S be able to get a permanent residence permit or her right to stay will be simply functional to the needs of the grandson? In other words, will she have to move once the grandson will not need her care anymore? Also, will G be able to get a work permit? These are all open questions that the Court, sooner or later, will have to face.

Overall, it seems that the Court is starting to endorse a strict approach toward free movement cases involving residence rights of third country nationals. This has probably taken many by surprise since, after Metock, the Court really seemed to have moved the pendulum toward a more family friendly and liberal approach. Once again, it will be interesting to see how and if this new and surprisingly strict trend will be upheld by new judgments in the future.

 Barnard & Peers: chapter 13

Thursday 27 March 2014

Can unemployed EU citizens be expelled and banned from re-entry?

Steve Peers

As detailed in today’s Open Europe blog post, the idea of expelling unemployed EU citizens has recently been mooted in Germany, and is therefore interesting the critics of EU free movement in the UK. It is therefore a good time to examine in detail in exactly what circumstances EU citizens can be expelled.

The main source of law on this issue is the EU citizens’ Directive, although the primary Treaty rules on the free movement of EU citizens play some role. While the Directive allows Member States to expel citizens on grounds of public policy, public security or public health, this possibility is irrelevant as regards the expulsion of unemployed EU citizens, since the Directive specifies that these grounds cannot be ‘invoked to serve economic ends’ (Article 27(1)). So the rules on expelling EU citizens on grounds of unemployment must be found elsewhere in the Directive.

Expulsion of EU citizens

First of all, it isn’t possible to expel EU citizens or their family members on the grounds of unemployment if they become ‘permanent residents’ in another Member State. This status is gained after five years’ legal residence in the host State (Article 16), or earlier if their work is interrupted due to retirement or permanent incapacity to work (Article 17).

Before that status is obtained, there are two phases in an EU citizen’s residence in another Member State. For the first three months of residence, an EU citizen can stay in another Member State without any conditions or formalities besides showing a residence card or passport (Article 6). However, there is no right to social assistance during that period (Article 24(2)), and EU citizens will lose the right to reside during this period if they ‘become an unreasonable burden on the social assistance system’ of the host State (Article 14(1)).

After the first three months, but before obtaining permanent residence status, an EU citizen’s right to reside in another Member State is conditional. The EU citizen must either: be a worker or self-employed person; have sufficient resources ‘not to become a burden on the social assistance system’; be a post-secondary student who makes a declaration concerning such sufficient resources; or be a family member of an EU citizen satisfying one of the first three conditions (Article 7). EU citizens retain the right of residence as long as they continue to meet these conditions (Article 14(2)). Conversely, they can lose the right to reside if they do not (or no longer) meet these conditions, but the Directive specifies that expulsion ‘shall not be the automatic consequence’ of an EU’s citizen’s recourse to the social assistance system of the host Member State’ (Article 14(3)).

Furthermore, it is not possible to expel a worker or a self-employed person, or a person seeking work, as long as such a person can ‘provide evidence’ that he or she ‘is continuing to seek employment’ and has ‘a genuine chance of being engaged’ (Article 14(4)). Any EU citizens subject to expulsion have the same procedural rights as those being expelled on grounds of public policy, public security or public health (Article 15(1)).

 Three other relevant rules in the Directive are important. First, EU citizens retain the status of worker of self-employed person (and so protection against expulsion on grounds of unemployment) if they are: (a) temporarily unable to work due to sickness or accident; (b) involuntarily unemployed, if they are registered with an unemployment office, although if they have been employed for less than one year, this status might terminate after six months; or (c) if they have started vocational training (a concept broadly defined by EU law), although this must relate to their previous employment unless they are involuntarily unemployed (Article 7(3)).

 Second, Member States cannot set a fixed amount to be considered ‘sufficient resources’, but must consider ‘the personal situation’ of each individual EU citizen (Article 8(4)). Third, if EU citizens die, leave or get divorced, their children retain rights to stay, as long as they are enrolled in education, along with the parent caring for them, even that parent is unemployed and reliant upon the social insurance system (see Article 12(3) of the citizens’ Directive and Article 10 of the Regulation on free movement of workers, as interpreted by the CJEU).

 So, to sum up these rules, which EU citizens can – and cannot – be expelled due to unemployment?

EU citizens cannot be expelled due to unemployment if they: (a) are permanent residents; (b) are workers or self-employed persons; (c) have ‘sufficient resources’; (d) have worked for more than one year and registered as a job-seeker; (e) have worked for less than one year and registered as a job-seeker, for at least a six-month grace period; (f) entered as a work-seeker, and can prove that they are still seeking employment and have a genuine chance of getting a job; (g) are family members of an EU citizen who has the right of residence or permanent residence; or (h) are the parent carers of an EU citizen’s child as described above.

Conversely, EU citizens can be expelled due to unemployment if they are not permanent residents, lack ‘sufficient resources’, and have become an unreasonable burden on the social assistance system, provided that they: (a) are not workers or self-employed persons, or the family members of an EU citizen or the parent carers of an EU citizen’s child as described above; (b) have worked for less than one year, if their grace period of least six months has run out; or (c) entered as a work-seeker, and cannot prove that they are still seeking employment and have a genuine chance of getting a job. Even if these criteria are satisfied, the legislation and case law make clear that each individual case must be considered before expulsion.

So there are small categories of EU citizens who can be expelled due to unemployment, but this is subject to an individual examination and procedural rights. Member States with a more liberal treatment of unemployed EU citizens than the Directive requires can tighten those rules, as long as they do not go so far as to breach the Directive. Clearly a blanket rule providing for the automatic expulsion of anyone unemployed for a particular period would therefore not be compatible with EU law.

Entry bans for unemployed EU citizens 

While the rules on the expulsion of unemployed EU citizens are complex, the rules on entry bans of the unemployed are very simple indeed. Article 15(3) of the citizens’ Directive states clearly that ‘The host Member State may not impose a ban on entry in the context of an expulsion decision’ that was taken on the basis of ‘grounds other than public policy, public security or public health’, ie on grounds of unemployment and poverty.

In light of this provision, it is remarkable that both the German and British governments are considering an entry ban of some sort. Are those plans compatible with EU law? The German intention is to impose an entry ban only on persons who have committed some sort of fraud. This raises the question of the correct interpretation of the ‘abuse of rights’ clause in the Directive (Article 35), which has yet to be clarified by the CJEU.

David Cameron’s proposal is that anyone who is expelled for not working would be subject to an entry ban for 12 months, unless they could prove that they have a good reason to return, such as work. On the face of it, this reason for an entry ban does not fall within the scope of the ‘abuse of rights’ clause, but rather within the scope of the rules which deal with the expulsion of those who are not working and who create an unreasonable burden for the social assistance system – where the Directive clearly rules out any entry ban.


While the Directive does allow EU citizens to be expelled due to unemployment and poverty, this is subject to tight substantive constraints, procedural rights for the persons concerned, and a case-by-case analysis. Moreover, a subsequent entry ban is ruled out. Both the German and British plans fall foul of these rules to an extent, the German plan because, at first sight, it sounds as if expulsion would be automatic, and the British plan because it would provide for an entry ban. Whether these rules should (or could) be changed is an entirely different question.

Barnard & Peers: chapter 13, chapter 16

When can Internet service providers be required to block access to websites?

Steve Peers

First of all, a confession. Until very recently, I had a sneaking (and very unfashionable) sympathy for the holders of intellectual property rights who sought to enforce those rights through heavy-handed injunctions against Internet users or service providers. After all, downloading music et al without paying for it is not really any different from shoplifting a CD. But then the BBC, having announced that the second series of The Bridge would be available on Iplayer for two weeks, curtailed that without warning to one week – leaving me unable to watch the final 20 minutes of the final episode of that series. (Note: this is not an invitation to send me spoilers.) I now have rather more sympathy for those who disdain copyright protection.

This brings us to today’s judgment of the CJEU in UPC Telekabel Wien, the latest ruling of the Court addressing the tension between the interests of copyright holders, on the one hand, and the telecom industry, on the other. The starting point is the EU’s Directive on copyright protection in the information society, which provides that:

‘Member States shall ensure that rightholders are in a position to apply for an injunction against intermediaries whose services are used by a third party to infringe a copyright or related right.’

In this case, two film companies sought an injunction against UPC Telekabel Wien, an Internet service provider, to block access by its customers to a website which offered streaming of the copyright product of the film companies. The Austrian court sent questions to the CJEU to establish whether the relevant rule in the copyright Directive applied in this case, and if so whether the EU’s Charter of Rights prevented the injunction being granted.

This case is the latest in a long line of CJEU judgments on similar issues. First of all, in its 2008 judgment in Promusicae, the Court ruled that EU law did not require Member States to insist that national courts be able to order injunctions against Internet service providers to keep records of their customers’ access to copyright material. But nor did EU law prohibit Member States from doing so. If Member States took up this option, EU law required them to balance the property rights of the right-holders with the data protection rights of individual users of the Internet.

Subsequently, the Court was asked to rule on exactly where that balance lay. In Scarlet Extended, it ruled that it would be a disproportionate breach of the Charter rights to the privacy and freedom of expression of Internet users, and the right to carry on a business of service providers, to order an injunction against the latter to keep records of users’ Internet use indefinitely. However, in Bonnier Audio, it ruled that a much more targeted measure, focussing only keeping records of the users’ access to a particular website where those users were downloading copyright material on a large scale, struck the right balance.

 In today’s judgment, the data protection and privacy rights of the Internet users were not at issue, because there was no element of data retention raised by the proceedings. The Internet users would (if the injunction were granted) simply going to be blocked from accessing the copyright material. So the Charter rights in question were the property rights of the right-holders, on the one hand, and the right to carry on a business of the service provider, and the users’ freedom to receive information, on the other.

The Court first of all ruled that Internet service providers were ‘intermediaries’ for the purposes of the information services copyright Directive. Therefore the right-holders were entitled to apply for an injunction against them. This approach makes sense, as right-holders would hardly be able to enforce their rights in practice if they could not obtain injunctions against service providers.

Next, the Court turned to the Charter issues. In doing so, it provided some useful clarification of the Charter right to carry on a business, which ‘includes, inter alia, the right for any business to be able to freely use, within the limits of its liability for its own acts, the economic, technical and financial resources available to it’. The injunction infringed that right because it ‘constrains its addressee in a manner which restricts the free use of the resources at his disposal because it obliges him to take measures which may represent a significant cost for him, have a considerable impact on the organisation of his activities or require difficult and complex technical solutions’. However, it did not strike at the ‘very substance’ of the right, since it left the service provider free to take different measures and to avoid liability by showing that its steps were reasonable.

As for the freedom of expression, the service provider had to ensure that it ‘strictly targeted’ the actions it took, to ensure that Internet users could still access any information which they had a lawful right to view. Furthermore, in order to enforce that Charter right, Internet users must also have a procedural right to challenge any measure taken by the service provider. However, the service provider nevertheless had to do as much as it reasonably could to limit access to the copyright material.


The novelty of this case is that there are not just two, but three rights at issue, each held by a different party: the rights holder, the service provider and the Internet users. The service provider is the ‘piggy in the middle’, having to ensure protection of the right-holders’ property rights, as well as the users’ freedom of expression rights, while still being entitled to invoke its own right to carry on a business. While the concept of the right to carry on a business is very broadly defined here, that is counter-balanced by the Court’s reinvocation of its traditional case law making it easy to justify interferences with that right.

Overall, a reasonable balance between the three conflicting rights in this case has been struck by the Court. If it is technically possible to reconcile users’ access to legal information, the service providers’ business interests and the intellectual property rights of the right-holders, this is the ideal solution. Moreover, since a solution of this sort does not involve data retention, it does not infringe upon privacy rights and so it is not possible (or necessary) to bring proceedings against users. In fact, surely data retention is disproportionate when such highly targeted blocking measures can be taken instead. The judgment may not be satisfactory from the perspective of those who dislike the very existence of copyright law, but I still believe that the creators and publishers of music, film and books et al are entitled to a reward for their effort.

In the case of The Bridge, of course, that reward has already been provided for, by means of the payment of British TV-owners’ substantial licence fees to the BBC. So if I happened to download those final 20 minutes of the final episode inadvertently, I would obviously not be betraying my principles.

Barnard & Peers: chapter 9

Monday 24 March 2014

Reverse discrimination against rape victims: a disappointing ruling of the CJEU

Steve Peers

While national criminal justice systems are often criticised for ignoring the role of crime victims, the EU has shown great interest in this issue from an early stage of the development of EU criminal law. One of the first EU 'Framework Decisions' concerned crime victims' rights, and this measure has since been replaced by a comprehensive Directive on this issue, which Member States must apply by October 2015.

Yet this Directive is not the only measure dealing with crime victims, as the topic of state compensation for crime victims is the subject of an earlier Directive, dating from 2004. The primary aim of the compensation Directive is to establish a framework for compensation in cross-border cases, where a 'violent intentional crime' is committed in a Member State other than the Member State where the person concerned is habitually resident. However, the Directive does provide that 'All Member States shall ensure that their national rules provide for the existence of a scheme on compensation to victims of violent intentional crimes committed in their respective territories, which guarantees fair and appropriate compensation to victims.'

In the recent order in C, a woman in Italy brought a claim for state compensation on the basis of this provision, following the conviction of her attacker for sexual violence. The national court had ordered the offender to pay compensation (an issue addressed by the general Framework Decision, now the Directive, on victims' rights), but there was no prospect of enforcing this order, since he did not have the money to pay that compensation. So she tried to obtain compensation from the Italian state instead, pursuant to the Directive. However, according to the CJEU, the Directive did not apply to purely domestic cases like this one, but only to cases where there was a cross-border element.

With great respect, this is just not convincing. While the main focus of the Directive is certainly compensation in cross-border cases, Article 12(1) of the Directive makes clear that this takes place on the basis of each national system for compensation. Therefore Article 12(2) - quoted above - requires each Member State to set up a national system covering crimes like this one. So a failure by Italy to provide for state compensation for its residents who are victims of such crimes will complicate any attempt by visitors from other Member States to collect compensation from the state in such cases.

Since such cross-border disputes will undoubtedly fall within the scope of the Directive, the CJEU might be prepared to rule that crime victims in such case would nonetheless have a directly effective right to compensation against the Italian government, regardless of its failure to implement the Directive fully. While protection of some victims is of course welcome, all of them ought to be protected. EU lawyers have long become accustomed to 'reverse discrimination' as regards family reunion, where EU citizens who move between Member States can rely on decent standards, while those who do not leave their Member State can in some countries (like the UK) be subject to a far more restrictive national law. It hardly helps the legitimacy of EU law to accept that the same double standards apply to compensation of victims of violent crime, especially when the better interpretation of the legislation concerned is that they shouldn't.

[update: see analysis of the main crime victims' Directive, which has applied since autumn 2015, here].

Barnard & Peers, chapter 25

Friday 21 March 2014

Agency workers: Are they sufficiently protected by EU law?

Steve Peers

'Atypical' workers (those workers who do not have a full-time permanent contract) are often in a very vulnerable position in the labour market, and so are possibly subject to exploitation. The use of such forms of employment might also undercut other workers. On the other hand, on some occasions such employment does suit the needs of both workers and employers, and constitutes an important form of job creation. To strike a balance between these two principles, the EU has adopted measures to regulate three specific categories of atypical workers: agency workers, part-time workers and fixed term workers.

The rules governing agency workers are set out in a Directive adopted in 2008, which Member States had to apply by December 2011. Today, the Commission released its report on the application of this Directive. To what extent is the Directive achieving its parallel objectives of ensuring that agency workers are protected, while also encouraging job creation?

Basic issues

First of all, the Directive does not apply to employers who are not conducting 'economic activities' (the Commission interprets this to mean employers not selling their service for remuneration). Also, Member States may exclude from its scope those workers in a special job creation scheme. The Commission reports that most Member States do not apply these exceptions; those which do have correctly applied them.

As for the definitions in the Directive, only one Member State has specified that the temporary work agency must be considered the employer. This is definitely the correct interpretation of the Directive, so it is unfortunate that no more Member States have made this clarification.

Worker protection

The key rule as regards employee protection is the principle of equal treatment for working conditions. Here the Commission has some doubts about some Member States' rules requiring a 'comparable employee', whereas the Directive refers to the 'same job'. But this form of transposition is surely fine as long as employers and (if necessary) courts recognise that a worker doing the same job in the user undertaking is always a comparable employee.

However, there are derogations as regards equal treatment. First of all, temporary workers might not be entitled to equal pay if they are paid by agency between assignments, and have a permanent contract with the agency. Only five Member States use this derogation. In several cases, workers are entitled to only half of their previous pay (subject to the national minimum wage as a floor). The Commission raises some doubts about this, but says only that it will hold further discussions.

Secondly, temporary workers can receive equal treatment pursuant to a collective agreement, if that agreement 'respect[s] the overall protection' of temporary workers. Ten Member States apply this rule, and the Commission does not believe there is any problem with its application. While it might generally be assumed that unions can defend the interests of workers, it is possible in principle that in some cases they have prioritised the interests of permanent workers instead, or are in a weak bargaining position and so have accepted terms that do not respect temporary workers' protection.

Thirdly, in States where collective agreements are not generally applicable, there can be a derogation from the principle of equal treatment, provided again that 'the overall protection' of temporary workers is respected. This might include a qualifying period, and might specify that employer pensions are not included within the scope of the equal treatment rule. Only the UK and Malta apply this rule. The UK provides for a 12-week waiting period, while Malta disapplies equal treatment for the first 4 weeks (if the assignment is longer than 14 weeks). Both Member States exclude employer pensions from the equal treatment rule.

The derogations are subject to the requirement that there must be national measures to prevent their 'misuse', and in particular to prevent successive assignments being used to circumvent them. As the Commission points out, this is a particular risk where there is a qualifying period for obtaining equal treatment. While the EU's fixed-term work Directive restricts the successive use of fixed-term contracts, the CJEU has ruled in the Della Rocca case that this Directive does not apply at all to temporary workers. So the question of what constitutes misuse must be defined from scratch. It is unfortunate that the drafters of the agency work Directive did not simply apply the fixed-term work Directive rules, in order to provide for some legal certainty. The Commission does not comment further on the rules used to prevent misuse, or on the Member States' use of the third derogation.

Job creation

The Directive requires Member States to review existing barriers to agency work, and states that such restrictions can only be justified in the general interest, in particular as regards health and safety, protection of abuse, worker protection and functioning of the labour market. According to the Commission, while Member States have reviewed their national law, few have changed it, justifying their existing restrictions on one of the grounds set out in the Directive.

However, it could be argued that the interests of worker protection and protection of abuse are already addressed in the Directive, and the issue of health and safety is already addressed by separate legislation (referred to in the report) which allows Member States to prohibit temporary workers from certain jobs on health and safety grounds. Some Member States have banned temporary workers from being used as 'scabs' to replace striking workers; this possibility is clearly authorised by the preamble to the Directive.

Other issues

The Commission finds no problem with Member States' application of the rules on workers' representatives, access to employment and dismisses the arguments of industry that the Directive imposes too many cost requirements for them.

Overall, the Commission concludes that there is no need to revise the Directive, and states its intention to hold further discussions about its correct interpretation. It should be noted that there is one case pending before the CJEU about this Directive: the AKT case, which asks the Court to clarify whether certain national barriers to the use of agency work can be justified. Furthermore, in its judgment in Michaeler, the CJEU held that national rules restricting part-time work breached the EU's part-time workers Directive, and had to be set aside by the national court. That judgment could obviously be relevant by analogy.


The Commission clearly does not wish to upset anyone by proposing the amendment of this Directive, or even as regards its application. It dismisses the objections of its critics rather summarily also. This is a classic case where a thorough 'impact assessment' needed to take place, not (as is usually the case) before legislation is proposed, but in order to determine the actual impact in practice. How much have workers benefited from this Directive, and at what cost to employers? Have jobs been created or lost?

Also, the Commission could usefully have more clearly indicated its view as to the correct interpretation of the Directive, for instance in a guidance document. It's hard to work out what the Commission actually thinks, besides discerning an unwillingness to disagree with anyone.  As noted above, there are good reasons to think that some national restrictions on the use of agency work might be open to challenge, and that some Member States are not sufficiently complying with the minimum requirements for agency workers' protection. The Commission's approach to the Directive in its report does not really help to clarify these matters.

[Update: the CJEU gave its ruling in the AKT case on agency workers in 2015. See analysis here.]
[See also: my analysis of the issue of equal treatment of atypical workers, in an article published in the Yearbook of European Law.]

Barnard & Peers, chapter 20

Tuesday 18 March 2014

Surrogate Mothers' Day in the Court of Justice: Reflections on EU sex equality law

Steve Peers

Women who would love to give birth to children, but who are unable to do so for medical reasons, have always understandably attracted great public sympathy. Historically, the only option for them was adoption. But in the last few decades, other options have become available, in particular IVF and surrogacy. 

While EU law does not regulate any of these issues as such, it does regulate maternity leave, by means of the pregnant workers' Directive. Moreover, some other aspects of the employment implications of pregnancy are addressed by the Directive on sex discrimination in employment. For instance, the latter Directive bans discrimination against a woman who is undergoing IVF treatment but is not yet carrying an unborn child, according to the judgment of the Court of Justice of the European Union (CJEU) in Mayr.

But what is the position of surrogate mothers as regards maternity leave? Today the CJEU addressed that issue for the first time, in the judgments in CD and Z. The CD case concerns a British woman who is the commissioning mother of a child who was genetically fathered by her partner, while the Z case concerns an Irish woman who is the commissioning mother of a child who is genetically hers and her husband's. In both cases, a surrogate mother carried and gave birth to the child.

The Court took the view that the pregnant workers' Directive, which sets out the right to maternity leave, can only apply where the same woman becomes pregnant, gives birth and then takes maternity leave as regards a newborn baby (or babies), applying the Mayr judgment, which stated that the Directive only applied once a person was pregnant. The sex discrimination directive could not apply either, because a commissioning father would be treated the same as a commissioning mother; there was no evidence of indirect discrimination; and its ban on discrimination against women who have been pregnant or gone on maternity leave could not apply, since the pregnant workers' Directive did not apply. Since the issue fell outside the scope of EU law, the EU Charter could not apply as regards assessing the validity of the sex discrimination Directive.

In the Z judgment, the CJEU reiterated its reasoning, and also added some further points. It noted that the sex discrimination Directive expressly states that adoption leave is an optional matter for Member States. It then answered questions about the framework equality Directive, notably about the ban on discrimination on grounds of disability in that Directive. In the Court's view, since the EU Directive only applied to discrimination relating to employment, the disability in question has to hinder the worker's participation in professional life. But Ms. Z's condition did not affect her access to employment. Finally, the Court ruled that the validity of the framework equality directive could not be reviewed due to incompatibility with the UN Convention on the Rights of Disabled Persons (which the EU has ratified), because that Convention was only 'programmatic'.

These two different cases were the subject of opinions by two different Advocates-General, who expressed two radically opposed viewpoints. The CJEU essentially followed the opinion of Advocate-General Wahl in the Z case. On the other hand, Advocate-General Kokott, giving the opinion in the CD case, took the view that in light of social developments and the EU Charter of Fundamental Rights, the pregnant workers' Directive was capable of applying to cases where one woman carried and gave birth to a child, while another woman then took care of the newborn baby. Inspired perhaps by the judgment of Solomon, she then suggested that the maternity leave should be split between the two women concerned.


It seems odd that women who undoubtedly consider themselves as being new mothers and who are indisputably looking after a baby who is genetically theirs (or their partner's) fall outside the scope of EU law, given its detailed regulation of maternity leave. Certainly, the key feature of these judgments - the intention that the pregnant workers' Directive apply to the same woman who got pregnant, gave birth and looks after a child - is surely a correct assessment of the legislature's intention. Having said that, the application of the Mayr judgment is a little unconvincing, since it concerned the question of whether the woman concerned had ever got pregnant at all, long before she had a baby to look after.

The EU legislature's decision to treat adoption leave differently from maternity leave and pregnancy in the sex discrimination Directive is undoubtedly clear too. And logically, since the framework equality Directive only applies to employment, it can only be applied to a disability linked to employment. But the CJEU's application of this principle is not fully convincing. In fact, the case is difficult to distinguish on this point from the Coleman judgment, where the disablity of the worker's child was in principle connected to her loss of employment, so the Directive applied. If the Directive applies where there is a link between the worker's employment and another person's disability, why does it not apply where there is a link between the worker's employment and the worker's own disability, as long as that disability has an impact, however indirect, upon her employment - in this case, her maternity leave?

As for the UN disability rights Convention, the Court's ruling that it cannot be relied upon to challenge the validity of EU law is, with great respect, clearly wrong. The Convention includes a ban on disability discrimination in its Article 5(2): this is exactly the sort of equality clause that the Court usually accepts can be directly effective, and which therefore provides for a ground for challenging the validity of EU acts. A better approach, leading to the same result, would have been to rule that the application of the Convention to the issues of the Z case falls within the competence of the Member States, not the EU (the Convention is a 'mixed agreement'), given the Court's ruling that EU legislation was not applicable to this case anyway.

There is, of course, direct discrimination in these cases: not between men and women (since men can also be infertile), but between different categories of mothers: those who are looking after a child whom they have given birth to, and those who are looking after a child whom they have not given birth to. The two groups are not similarly situated as regards pregnancy and childbirth, but they are in an identical position as regards childcare. Surely the general obligation to ensure equal treatment set out in Article 21 of the Charter could have been relevant here.

Furthermore, the Court's assertions as regards indirect discrimination are not convincing, as it is surely more likely in practice that women, rather than men, will want to take time off work to look after a newborn. Moreover, in a case involving same-sex couples, it could be arguable that a failure to give them equal treatment with opposite-sex couples as regards maternity or adoption leave would be discriminatory on grounds of sexual orientation, given that it is not possible for same-sex couples to have a child who is genetically the child of both of them.  

The Court's literal approach to the EU legislation has also lost sight of its underlying purpose in two respects. First of all, a woman who wishes to take time off work following the adoption of a child (whether there has been a surrogate mother involved or not) faces the same work/life balance issues as a woman who has given birth to a child. The underlying goal of EU law on sex equality as regards employment is to ensure that women have equal access in practice to the workforce. Women who have adopted a child are bound to face problems in that regard, similar to those facing women who have given birth. Secondly, from the perspective of the best interests of the child (an issue addressed by Advocate-General Kokott), and the objective of ensuring that a mother can bond with her child (referred to expressly by the Court), it should not matter whether that child has been adopted or not.

These points lead us to a more radical question: why should the EU legislature itself facilitate massive inequality as regards the care of newborn children? Advocate-General Kokott, noting that the Charter requires the child to have contact with both its parents, then applies that to the rights of the mother. But if we are going (as she suggests) to sever the link between pregnancy and childbirth, on the one hand, and childcare on the other, then why should we assume that the childcare has to be provided by the mother, by means of granting women maternity leave, but not granting parallel (or transferable) paternity leave to men? Indeed, from this perspective, the more generous the maternity leave, the greater the distinction between the roles of men and women as regards childcare, with a corresponding greater impact upon engagement in the workplace. Simply put, maternity leave is the gilded cage of sex discrimination.

Barnard & Peers: chapter 20